BRAZIL S TRANSNORDESTINA RAILWAY AS CRONY CAPITALISM:

Transcription

1 BRAZIL S TRANSNORDESTINA RAILWAY AS CRONY CAPITALISM: ABSTRACT Lessons from TCE, Public Choice Theory and Austrian Economics Vladimir Fernandes Maciel Roberta Muramatsu Maurício Fronzaglia Álvaro Alves de Moura Jr. This paper draws on insights from Transaction Cost Economics, Public Choice Theory and Austrian Economics to advance the thesis that government policies to foster economic growth of Brazilian northeast states through an ambitious investment project for the construction of Transnordestina railway connecting the states of Piaui, Ceará with Pernambuco and its Suape Port makes a case for government failures and manifestations of crony capitalism in an institutionalized rent-seeking society that carry negative consequences for economic efficiency and sustained growth. Keywords: crony capitalism, rent-seeking, public choice, Austrian economics, transaction costs JEL Classification: B5, B50, B53, D23 Introduction It is a commonplace to state that the Brazilian political as well as economic institutions have evolved with large incentives for rent-seeking activities and obvious manifestations of crony capitalism. This is partly so because the government has a long-standing tradition in which public officials and their federal, state and municipal agencies interact with private investors as major or minor shareholders and financial backers (Musacchio and Lazzarini 2014; Lisboa and Latiff 2012, Haber 2002). In a complementary line of reasoning, Paul Dragos Aligica and Vlad Tarko (2014) claim that crony capitalism in South American countries amount to a mechanism for securing property rights in the absence of rule of law by creating an overlap between those who have the power to confiscate property and the owners of large firms who have property that could be confiscated (p.157) This article is an attempt to scrutinize the items that explain one big Brazilian government failure regulation of the Transnordestina Railway that purports to promote regional economic development in the Northeast parts of Brazil and to lower its high transaction costs constraining the country s potential for international division of labor. More specifically, it draws on an approach that combines insights from Transaction Cost Economics, Public Choice Theory and Austrian Economics to assess whether, why and how exactly Transnordestina Railway is a predictable manifestation of crony capitalism in the Brazilian rent-seeking society (Williamson 1985, Tullock 2002, Cobin 2006, Ebeling 2016, Boettke and Edwards 2002, Benson 2002). Our main aim is to uncover how the contractual arrangements have changed over time and their unintended consequences in response to the public officials knowledge problem, their shifting incentive structures and actual decision-making procedures

2 underlying the very design and implementation of the Transnordestina railway project (Holcombe 2012). In addition, it is worthwhile to stress that the private shareholder of Transnordestina Logistics S.A (henceforth:tlsa) is Companhia Siderúrgica Nacional (CSN), a former state-owned National Steel Company privatized in the early 1990s. More precisely, this article focuses on the distorted relationship between the private railroad operator (Transnordestina Logistics S.A) and the public agencies playing the role of regulating, managing and funding transport infrastructure. We focus our attention on TLSA connections with BNDES (Brazilian National Bank of Economic and Social Development), VALEC Engenharia, Construções e Ferrovias S.A. (the public company responsible for planning the Brazilian railroad investments), BNB (Bank of Brazilian Northeast region), FINOR (Investment Fund for the Northeast of Brazil), SUDENE (Superintendence for the Economic Development of the Northeastern Brazil) and FDNE (Development Fund of the Brazilian Northeast). We go on to argue for the idea that the systematic government failures surrounding the unfinished Transnordestina Railway project provide a case study for testing our hypothesis that most manifestations of crony capitalism in Brazil are (mal)adaptations of an unstable business environment with concentrated benefits and disperse costs, where government officials shift their roles to benefit named private investors and promote misallocation of scarce resources. This paper is organized as follows. Section 1 describes the historical background of the Transnordestina railway project from its initial phase in the late 1960s until now. Based on a transaction cost economics approach, section 2 presents and analyzes the operation of Transnordestina railway as part of CSN s verticalization strategy that results from its hierarchical governance structure to minimize transaction costs and seek efficiency. Section 3 draws on a combined Austrian Public Choice approach to assess critically the implementation of Trasnordestina railway project and to explain why it makes a great case for addressing distorted contractual arrangements, the pervasive hold-up problem in a world of incomplete contracts with agents of disperse knowledge and the peculiar aspects of crony capitalism in a rent-seeking society. It goes on to tentatively present and discuss some troubling coordination strategies among bureaucrats from Ministry of Transportation, the Ministry of Domestic Affairs, VALEC, SUDENE, BNB and private actors coming from TLSA, CSN among other business associates. Section 4 wraps up the overall argument, briefly draws some implications and concludes. 1. Digging deeper into the institutional background of Transnordestina railway This section presents and discusses the institutional background in which negotiations and institutional arrangements to build up Transnordestina Railway have evolved. The railway investment project was part of the Economic Growth Acceleration Plan (PAC) 1 designed under the second term of ex-president Luiz Inácio Lula da Silva ( ). According to the government rhetoric, PAC purported to be a Keynesian-oriented set of public policies to minimize the economic downturn following the 2008 financial crisis, promote Brazilian economic growth and the country s increase social indicators. The official narrative was built on the premise that public incentives for investments in Transnordestina railway would allow for regional integration and sustained socioeconomic improvements in the Northeast areas of Brazil. 1 In this article, all the proper names of Brazilian and foreign entities, projects etc. were freely translated to English, followed by their abbreviations or acronyms as used in Portuguese.

3 Given the complexity of the issues at stake, we appeal to complementary highly explanatory perspectives Transaction Cost Economics, Public Choice Theory and Austrian Economics to assess more critically the puzzles surrounding Transnordestina investment project delays, changes in its institutional arrangements and increased noncredible budget requirements. More precisely, we claim that the related public action sheds light on episodes of government failures like rent-seeking in a distorted economic system called crony capitalism. Our main aim is to provide an analytical understanding of the institutional flaws underlying the ambitious project of Transnordestina Railway, which remains unfinished. The construction of the Transnortheastern Railway has gained much room in the Brazilian press and gave rise to several political arrangements and conflicts. To the best of our knowledge, making sense of the ongoing delays and the increasing budgetary requirements depends on a clearer account of how private and public actors with disperse knowledge make choices in a weak institutional environment where crony relations do pay off highly to the detriment of economic efficiency and productivity gains. The idea of constructing a railway to link the hinterland of the Brazilian northeast region with the coast and its main ports dates back the 19 th century. Given Brazil s continental dimension, integration had always been a hot topic in the political as well as economic agendas. In late 1960s, the military government ( ) started to build a railway similar to the current Transnordestina project but it was soon put aside by public authorities and the works were interrupted soon afterwards and were resumed only in the next century, as displayed in Figure 1. Late 1960s Proposal of the Transnortheastern project 1997 CSN obtained the federal concession of Transnordestina railway network 2006 CSN presented to the government a new Project for Transnordestina railway 2007 Attempt to define property rights with details about the government s participation in financing Transnordestina 2014 Construction was interrupted due to dead-lock with Odebrecht developer Third contract amendment due to mispredicted highercosts. Contractual conflicts with TLSA developer in charge of undertaking the railroad construction and construction engineering agents with high political bargaining power Conflicts during contruction between local population and different railroad players. Second step of changes in contractural arrangements due to higher costs Beginning of railroad construction with first contractual amendments in response to increased costs of Transnordestina project 2015 New TLSA management staff comprised of former ministers, governors and stateowned bank executives 2016 Expansion of the share of State s resources in Transnordestina railway through VALEC (public company) interventions 2016 The Federal Court of Accounts (TCU) prohibited new financial contributions to TLSA and declared TLSA bankrupt. But the Court reversed its decision 2017 TLSA claim for additional Federal resources and The government tries to take back the concession of the railway

4 Figure 1 Timeline of the relevant facts underlying Transnordestina railway project In 1997, under a political and economic context that rewarded privatizations, the bidding process of the northeast part of the Federal Railway Network ended with great market enthusiasm. At that time, the bidding process winning group was CSN (National Steel Company). The government ascribed to CSN the right to explore all activities related to freight car transportation with Transnordestina railway (see Map 1). Surprisingly enough, the government officials never requested any previous economic feasibility analysis of Transnordestina enterprise to help them deal with their knowledge problem and designing better agreements that would specify clauses regarding budget constraints, and investment schedules, construction deadlines and punishments in case there were any opportunistic move of the contracting parts. The only piece of information referred to the fact that the federal government would take responsibility for the Transnordestina railway by means of financial provisions. Map 1 Plan of the New Transnordestina Railway Source: Brasil, Ministério dos Transportes (2013). In 2006 (at the end of President Lula s first term) CSN submitted to the federal government a R$8.0 billion worth project for the construction of 1,728 kilometers of the Transnordestina Railway. CSN also received alternative funding schemes for infrastructure projects, by means of a project finance with BNDES support (Ministry of Transport, 2005). 2 2 BNDES (National Bank of Economic Development) defines this kind of partnership as follows: Project Finance is a form of financial engineering supported contractually by the cash flow of a project, having as guarantee the assets and receivables of this same project. Available at

5 After several rounds of negotiations, Transnordestina project cost was revised and it lowered by R$4.5 billion in response to government pressures. According to the new proposal, the new company Transnordestina Logistics (TLSA), controlled by CSN, would be in charge of the construction and future railway operation. According to the investment project, the railway would be completed in three years. In 2007, the funding sources for the construction of the new railway were defined, as seen in Chart 1. TLSA would get loans of R$2.3 billion with the Northeast Development Fund (FDNE), and R$0.4 billion with the National Bank of Economic and Social Development (BNDES). The remaining sum would be invested in the form of TLSA capital increase by its controller CSN and the Investment Fund for the Northeast (FINOR). Therefore, the participation of the direct and indirect public administration of the Transnordestina Railway would total 76.4% of the scheduled investments. Chart 1 Sources of funding the Transnordestina project in 2007 (in R$ billion and %total) Source: Based on Valor (2013). After the actual beginning of the construction work in 2008, there was pressure for TLSA to require budgetary adjustments to the Transnordestina project due to big increases in labor and materials costs, such as concrete, steel and oil. The renegotiations with the federal government resulted in a new R$5.4 bi budget, distributed as seen in Chart 2. The FNDE increased its financial participation through loans and CSN made new capital increases. In addition, two federal state-run companies provided additional resources for the railway: the Engineering, Constructions and Railways (VALEC), which became the TLSA shareholder, and the Northeastern Bank (BNB). The latter promised to provide R$180 million with funds for Transnordestina railway. Due to the changes in the financial arrangements, the new deadline for completing Transnordestina construction would be 2014, the last year of ex- President Dilma Rousseff's first term. < Finance/>, retrieved on 20/06/14.

6 Chart 2 Composition of the resources of the project in 2008 (in R$ billion) Source: Based on Valor (2013). At the end of 2011, there was a demand for a new budget adjustment to the Transnordestina railway project. TLSA submitted the request to the to the federal government. After a few months of negotiation, TLSA players and government officials agreed on a R$1.3 bi increase in its budget schedule, which implied a total budget R$6.7 billion for Transnordestina project in The construction deadline was postponed for December Interestingly enough the last stages of negotiations made all the contracting parties - public funding sources and CSN, to increase its expenses in the Transnordestina Railway project, as shown in Chart 3.

7 BNB 2% CSN 10% FNDE 37% VALEC 32% BNDES 7% FINOR 12% FNDE FINOR BNDES VALEC BNB CSN Chart 3 Composition of the resources of the project in 2015 Source: Based on Valor (2015). In 2013, the Transnordestina Logistics (TLSA) again submitted to the Ministry of Transport a new proposal of budget revision with new deadline for completing Transnordestina railway. After many rounds of negotiations, the Federal government accepted a new agreement. The latter allowed for increased investments in Transnordestina by R$7.5 billion. Furthermore, the new agreed deadline would be September 2006 (see budget evolution in Chart 4). Under the new terms of the new agreement it is expected a R$400 million increase due to VALEC participation in the TLSA capital. According to recent CSN briefs, R$4.1 billion were invested in the project until 2013, around 53% of the total initial money investment schedule (Valor, 2014).

8 13,1 11,2 11,2 8 4,5 5,4 5,4 5,4 6,7 6,7 7,5 7, Original Approved Budget made budget for by TLSA the Federal Government First revision of the budget (increased costs) Second revision of the budget (increased costs) Third revision of the budget (increased costs) Promissed new budget due to TLSA and Federal Government contributions Estimated total budget needed (with additional Federal Government contribution) Chart 4 Evolution of the New Transnortheastern Railway budgets from its conception until the last contract revision (in R$ billion) 100,0% 90,0% 80,0% 70,0% 60,0% 50,0% 40,0% 30,0% 20,0% 10,0% 0,0% FEDERAL GOVERNMENT PRIVATE SHAREHOLDER The status of the Transnordestina project implementation and construction, displayed in Table 1, suggests that only three out of the 29 lots were fully completed during the period. This means that 216km out of 1.728km of Trasnordestina railway was ended, which corresponds to only 12.5% of the total project. It seems that there will still a long way to go before Transnordestina railway becomes reality.

9 Table 1 Indexes of the Transnordestina Railway execution in late 2013 Lots and extension Total investment Completed extension 29 lots and km Pernambuco: 743 km (10 lots) Piauí: 392 km (7 lots) Ceará: 593 km (12 lots) Source: Brasil, Ministério do Trabalho (2013). Total evolution R$ 7.5 billion 216 km 40% Salgueiro Missão Velha (96 km) 02 lots in Salgueiro Suape (120km) Infrastructure: 48% Works of Art Special: 41% Superstructure: 24% In 2015 the updated amount of the total investment was approximately R$9 billion with expectations of increased by R$11 billion. According to TLSA president, Ciro Gomes the Transnordestina railway might be ready in However, further complications and delays are expected to the currently turbulent political and economic environments, despite some federal government announces of public incentives for the Brazilian infrastructure sectors just like Dilma Roussef s II PIL (Second Program of Investment and Logistics) in In February 2015 reports coming from TCU (Federal Court of Accounts) encountered many irregularities in contractual relationships involving Transnordestina railway construction process. Some of the mistakes could imply cancellation of the railway concession contract and severe punishments to CSN. That sheds extra light on the fact that bureaucrats cannot overcome their severe knowledge problem and come to grips with government capture and crony relations (Valor, ). The Federal Court of Accounts (TCU, 2014) reported irregularities in government managing contractual arrangements and monitoring entrepreneurial goals. To complicate matters, there is no coordination between budgetary issues and investment schedules. In addition, many terms and conditions underlying Transnordestina concession contract failed to be followed with ANTT consent (Ground Transportation Regulatory Agency), which suggests some room for another government failures named regulatory capture. TCU reports go on concluding that TLSA has not complied with the deadlines. It has not respected budget constraints over time, either. In the beginning of 2016, VALEC, a run-state company attempted to promote Transnordestina railway construction of the railway by providing an additional funding of R$ 420 million. In response to the puzzles, the government decided to increase its share in the total capital of Transnordestina railway project. It began with a percentage of 8.2% and some politicians want the public agency VALEC to acquire 49% of TLSA. (VALUE, ). In the first half of 2016 two events involving TLSA seem to put the railway project under great threat. First, the TLSA was declared bankrupt (VALUE, ). Second, The

10 Federal Court of Accounts prohibited the federal government to make any public money transfer to Transnordestina (VALUE, ). With all this in mind, there seem to be no big prospects for resolving such anomaly. Transnordestina Railway may remain an unfinished infrastructure project due to government failures and an institutional environment of crony capitalism that rewards rent-seeking activities. 2.Making sense of Transnordestina railway through Williamson s economic lenses of transaction costs In this section, we briefly present and scrutinize Williamson s (1979, 1985, 1988, 2005) transaction cost economics (TCE) in order to understand the institutional arrangements and organizational forms underlying Transnordestina Railway. It is important to stress that TCE is one of the various strands of New Institutional Economics (NIE) that draws on Ronald Coase s (1937) view of economic organizations and their arrangements as initiatives to respond to positive transaction costs and achieve efficiency. According to Peter Klein s TCE portrait, TCE represents another approach to studying institutional arrangements. Here the emphasis is on governing transactions. TCE holds that all but the simplest transactions require some kind of mechanism what Williamson (1985) calls a governance structure to protect transacting parties from various hazards associated with exchange (Klein 1998, p. 13). More precisely, TCE regards organizations as a complex network of contractual relationships that purport to align incentives for cooperation. In Williamson s perspective, the very choice over a governance structure market, hybrid and hierarchy depends on behavioral parameters (opportunism and bounded rationality) as well as specific features of business transaction (frequency, uncertainty and asset specificity). The TCE framework stresses that agents are boundedly rational, i.e. they have constrained computational capabilities to anticipate all the contingencies that the future will bring. As a result, contracts have inevitably incomplete clauses that in turn bring extra risk to transacting parties. This is partly so because (mal)adaptation costs tend to be positive. The holdup problem is one empirically grounded example of maladaptation. Businesses learn that their relationship-dependent investments cannot be free from the risk of changing circumstances in the future that might motivate one of the trading parts to be willing to behave opportunistically. In this paper, we argue that the holdup problem seems to be a pervasive issue of the distorted Transnordestina railway arrangements. Following Klein, Crawford and Alchian (1978), specialized assets or investments that are not redeployable motivate business partners to try to expropriate quasi-rents and make strategic use of their personal relations with government officials and agencies to achieve their goals. Furthermore, we go on to examine whether uncertainties of the institutional environment amplify the incentives for contract breach, rent-seeking, a spiral of control and pressure for hierarchical organization. It is important to stress that site asset specificity and dedicated assets play a non-negligible role in explaining Transnordestina railway institutional arrangement, which is closer to a hierarchical governance mode than a market structure. To the best of our knowledge, the choice over building up a railway in the northeast of Brazil reveals an effort to economize

11 on transaction costs associated with transportation and regional integration, which put severe constraints on the country s productivity and competitiveness. 2.1 History matters for understanding the logic of Transnordestina governance structure and its contractual practices In 1997, Companhia Siderúrgica Nacional (CSN, National Steel Company) deliberately joined the former Northeast Railway Network bidding process and turned out to be the winner of this specific railway concession. Buying valuable logistic assets seems to be part of CSN s verticalization strategy to lower its high transaction costs and become less dependent on Companhia Vale do Rio Doce (CVRD, currently named VALE), the biggest Brazilian mining company with the most productive railway concessions. This is partly so because CSN deals with very high asset specificities or relationship-specific negotiations and investments that are indispensable for the very operation of its production line, which in turn depends crucially on an ongoing supply of iron ore. CSN s iron purchases require contractually defined prices and must respect their strategic plans and production schedules. It comes as no surprise that such requirements may call for a long-term relationship that creates incentives for cooperation and accommodating specific investments and its related sunk costs (Klein, Crawford and Alchian 1978). In tune with Williamson s (1985) line of reasoning, any contract is imperfect and would be too risky for CSN to fully depend on CVRD/ VALE, which controls most of the iron ore production and presents itself as the main supplier for the Brazilian steel industry (Vale has more than 80% of the iron production). 3 This leads us to suggest that CSN s organizational form can be regarded as an attempt to overcome high transaction costs and to respond to specificities of the institutional environment. To put it differently, CSN chose to pursue efficiency by internal (vertical) production, coordinating, under the same corporation, activities that would involve some contractual hazards if operated by several independent organizations in the market. In other words, CSN has verticalized its operations and started to extract its own iron ore since the 1990s through acquisition of Casa de Pedra, Engenho and Fernandinho mines in the State of Minas Gerais. It also engaged in transportation and logistics by becoming an important shareholder of the MRS Logistics S.A. The latter is entitled to operate important railways to transport the iron ore production in Minas Gerais state. MRS has administrative control over two terminals in Itaguaí Port in Rio de Janeiro state. 4 The foregoing institutional arrangement made CSN the second largest Brazilian iron ore producer, responsible for 4.59% of the total production in It is important to stress that CSN is aware of the fact that Brazil s state of Piauí has the fourth largest iron ore reserve, which remains unexplored. The economic prospects of mining activities in Piauí, where the final stretch of the Transnordestina railway purports to connect partly explains CSN s participation in the Rede Ferroviária do Nordeste (Northeast Railway Network) bidding process in It is important to stress that in the same period, iron ore reserves in Paulistânia city in Piauí started to be explored. Licenses for mining activities and exploration property rights Brazilian Mineral Yearbook published by the National Department of Mineral Production (DNPM), comprising the total amount of the production of Vale S.A. and its controlled company Brazil United Mining S.A. (MBR). 4 Available at < Brazilian Mineral Yearbook published by the National Department of Mineral Production (DNPM).

12 were exchanged. Under the so-called Planato Piauí Project, Bemisa (Brazilian Mineral Exploration Group) took the lead to structure the iron ore mining business in the region. Yet the effective mining operation depends on addressing the logistic bottleneck that could be somehow dealt by investments in Transnordestina railway. One of the main investors of the Planalto Piauí Project is the Opportunity Group, largely involved in several crony relations to be discussed later in this article. 3.Austrian-Public Choice insights to understand Transnordestina railroad project as government failure and expression of crony capitalism 3.1 Integrating Public Choice Theory with Austrian Economics Public choice is a framework that draws on insights of economics to account for the behavior of individuals with respect to government (Tullock et al 2002, p. 3). More specifically it assumes that people are people and therefore are actuated by their own self-interest. This implies the view that individual voters, politicians, regulators are just like voters and consumers who exhibit different behavioral outcomes because of the incentives they have under contexts of public and private choices. In Phillip Booth s words: At one level, Public Choice economics simply asks us to make the same assumptions about human behavior in the political sphere as we make when we analyze markets ( ) The self-interest operating in the political system will lead to government failure which can be more serious than market failure because of the coercive power that government exercises and because government is not subject to a direct competitive process. (Booth 2012, p. 12) The traditional literature of Public Choice Theory has the merit of representing government officials behavior in a less idealized way since they are just like other individuals pursuing in their interests. Due to the asymmetric information pervading the political spheres and voters rational ignorance, government failures (corruption, rentseeking activities and regulatory capture) are phenomena predictable by the economic approach to politics, Public Choice. In response to that, political economics propose transparency and accountability for self-interested politicians and regulators to alter their incentives to behave cooperatively and avoid cheating voters. In a nutshell, one way to control opportunism among public officials is to engage in disclosure of information and to establish a punishment to discourage some patterns of behavior. According to Boetkke and Lopez (2002), Hayek already addressed issues of public choice in his Constitution of Liberty and his Road to Serfdom. Before him, Mises already made clear his view that even if public officials were driven by public interest (other-regarding preferences), they cannot guarantee Pareto optimal allocations since they do not have the required complex pieces of disperse information about all the possible alternative uses of scarce resources. TCE, Public Choice Theory and Austrian Economics fit very well together to the extent that it sheds extra light on the fact that institutional environments with public choices involving concentrated benefits and disperse costs allow for some coalitions between named private agents with public officials interested in (political as well as money) profit opportunities, which in turn result in economically inefficient contracts and policies. In this perspective, government players are also entrepreneurs trying to obtain political and economic gains from their capacity to assign property rights to some private groups. Of

13 course, such patterns of crony relations depend on whether political and economic institutions of the country under study are exclusive (Acemoglu and Robinson, 2002). Additionally, incorporating Austrian insights into a public choice account of Transnordestina allows us to understand that part of government failures that call for new stages of intervention, regulation and capture might result from the empirical fact that public officials and members of agencies cannot get rid of the knowledge problem. In his discussion of on the "pretense of knowledge", Hayek (1989) stressed the confusion between practical and theoretical knowledge. Regulators and public officials tend to be overconfident about their knowledge regarding the functioning of complex phenomena. Following Hayek (1967), theoretical knowledge of complex phenomena can only be abstract, in the sense that cannot yield detailed predictions, but only pattern predictions of general characteristics of the object studied, the scope for successful regulation that required detailed information about dispersed information is severely reduced. This, in turn, suggests that nudge proposals should explicitly include the premise that regulators cannot obtain some knowledge about complex decision tasks with which individuals deal in the markets. Regulators tend to believe that they know better the problems with which the targets of the policies face. This is because public officials have more access to information and (non) experimental date. This view, sometimes endorsed by some Public Choice theorists underestimate subjectivity and some policy areas, regulatory schemes and even contractual relations that are emotionally charged. Such complications distort bureaucrats capacity to distinguish scientific knowledge and practical information. Some regulatory measures might bias bureaucrats attention and local knowledge in a way that they focus on goodness of intentions without evidence of their real effects and some disastrous unintended consequences. To complicate matters, public officials go on to try to correct government failures originating from unintended consequences of previous regulations. They tend to react with more regulations of the same nature instead of abandoning the original plan. This fits perfectly into the description of the dynamics of interventionism studied by Mises (Ikeda, 1996), in which appear the same conviction spirals in the regulatory process. Just like normal people who overestimate their understanding of complex phenomena and often appeal to superficial knowledge about patterns to draw conclusions about the nature/essence of a phenomenon, regulators and bureaucrats are not immune to the illusion that their knowledge is sufficient to identify some policy or regulation bottlenecks and to design institutional devices that help people behave as if they were fully rational. Such cognitive bias in the direction of expansion of regulation is reinforced in paternalistic institutional environments. It is likely that the regulatory apparatus might attract experts in favor of regulatory activity. Hayek (1979, p 290) describes accurately the positive feedback effect favoring expansion of regulation: The new kind of expert... is an expert in a particular institutional setup. The organizations we have created... have grown so complex that it takes more or less the whole of a person's time to master them. The institutional expert is not necessarily a person who knows all that is needed to enable him to judge the value of the institution, but frequently he is the only one who understands its organization fully and who therefore is indispensable.... [A]lmost invariably, this new kind of expert has one distinguishing characteristic: he is unhesitatingly in favor of the institutions on which he is expert. This is so not merely because only one who approves of the aims of the institution will have the interest and the patience to master the

14 details, but even more because such an effort would hardly be worth the while of anybody else: the views of anybody who is not prepared to accept the principles of the existing institutions are not likely to be taken seriously and will carry no weight in the discussions determining current policy. Hayek emphasizes two central elements: complexity and subjectivism. How is it possible increase the degree of complexity of structures if knowledge about its working details is limited? The answer to this question, present in all his theories, involves an evolutionary process of trial and error learning, which assumes both mechanisms of negative feedback and the initial freedom to undertake new solutions. It is important to stress that decentralized trial and error learning mechanisms are essential for the class of knowledge relevant to dealing with the coordination problem. It is not the abstract scientific knowledge, but the practical and dispersed knowledge of the agents about the "particular circumstances of time and place." To Hayek, the pretension of control of social process by central planners and regulators is due to confusion between these two forms of knowledge. As a result, policy makers, development experts, regulators among other bureaucrats had better be prudent to avoid the dangers of transferring the simplicity underlying their models to the complex social systems in the real world so as to legitimate increasingly centralized solutions and strong regulations. The lesson to be drawn is that effective government intervention is not possible because: a) knowledge is dispersed among individuals, b) it is subject to a continuous flow of change and c) knowledge refers to some unobservable factors. Failing to distinguish the foregoing categories giver room for never-ending spiral of corrective and interventionist actions. Under some institutional environments, such trend implies consolidation of rentseeking activities as expressions of crony capitalism Transnordestina railway and its unfinished investment projects: a case for government failure in an institutionalized rent-seeking society. We now turn to draw on the above insights from a perspective that integrates new institutional economics, public choice theory and Austrian ideas to understand why Transnordestina projects and contractual arrangements failed to be reinforced, which implied increasing costs and investment schedule delay. Our conjecture is that the recurrent pattern of concentrated benefits with disperse costs gives rise to rent-seeking moves and consolidation of crony relations, taken as maladapted outcomes. James Buchanan (1980) uses the term rent-seeking to describe behavior in institutional settings where individual efforts to maximize value generate social waste rather than social surplus. (1980, p. 47). In a similar fashion, Tollison (1982) appeals to rent-seeking to refer to the expenditure of scarce resources to capture an artificially created transfer (1982, p. 77). Quite recently, Lisboa and Latiff (2013) write a paper about democracy and growth in Brazil suggesting that rent-seeking is the process by which certain groups obtain privileges and benefits from government agencies (2012, p.1). The above portraits of rent seeking shed extra light on Randall Holcombe s view of crony capitalism s crony capitalism as an economic system in which the profitability of business depends on political connections that is certainly a by-product of big government (2013, p. 542). As consequence, there will always be some ambiguity surrounding the term crony capitalism. In response to that, we will favor Paul Dragos Aligica and Vlad Tarko s

15 (2014) view that crony capitalism is much more than an application of rent seeking perspective but also a mechanism for securing high rents to specific pressure groups due to their high political bargaining power. Our choice for the above interpretation of crony capitalism is based on our view that crony capitalism is South America amounts to a mechanism for securing property rights in the absence of rule of law by creating an overlap between those who have the power to confiscate property and the owners of large firms who have property that could be confiscated (p.157) To our minds, one big Brazilian government failure regulation of the Transnordestina Railway fits well with manifestation of crony capitalism and institutionalized rentseeking. In tune with the proposed Public-Austrian economic approach, all individuals are driven by their own self-interest and make use of information within the institutional environment to explore their entrepreneurial opportunities, even those that are not conducive to efficient allocation. Lisboa and Latiff (2013) put the issue very clearly: Contrary to mature democracies, several of the policies do not go through an encompassing government budget, but are conceded by means of specific government agencies. Decisions are taken in order to protect a specific group without taking care of total social costs. This is a distinctive feature of Brazil s political processes, with institutionalized agencies able to provide privileges and benefits to specific sectors ( ) without accountability of the costs (2013, p. 6) In order to examine our conjecture, we begin by assessing: a) CSN s participation in Northeast Railway Network bidding process, and b) features of the railway concession contract and agreements between CSN and government officials to build the Transnordestina railway. Our hunch is that the presence of the CSN, a strong player with several political relations and connections on all levels of government power has influenced the way the institutional arrangement has evolved over time. This is also due to issues of path dependence, since CSN was a run-state business until its privatization in The Northeast Railway Network concession contract in 1997 granted the operation of the former São Luís-Teresina Railway and the Teresina-Fortaleza Railway to CSN. In order to foster construction of new railway branch lines like Transnordestina the federal government promised to give incentives for CSN that would be entitled to explore the potentially high gains from rail freight and cargo transport operations in the future. In response to long delays and obstacles to advances in Transnordestina project, the federal government negotiated changes in institutional arrangement with CSN. Unlike the prevailing contractual relationship of the late 1990s, a new company Transnordestina Logistics S.A (TLSA) was created and the cash flow of its future operations would be provided as loan guarantee. Such contractual change purported to be an extra incentive that the government gave to speed up the Transnordestina project. In this setting property rights were restructured through two alternative options of legal channels orchestrated by the government. First, the federal government, probably through its state-run engineering company VALEC, would be responsible for the implementation of Transnordestina railway. As soon as the project is completed, Transnordestina railway would be granted to CSN (the private operator). Second, federal government proposed to renegotiate the contract with CSN by revoking the previously agreed concession clause and establishing a new bidding process. The latter would result either in a Public-Private Partnership (PPP), wherein a private company would be responsible for building, operating and

16 managing the railway. The federal government would in its turn guarantee a minimum profitability tax rate or grant the railway operation rights to a private company. However, the federal government soon realized the negative consequences of its renegotiation moves with CSN that could imply reputational losses and decreases in government credibility resulting from contractual breach. The solution public officials at the end proposed involved creation of a new company TLSA jointly run by CSN and the federal government under subtle financial engineer devices of project finance. Under this new setting, the government presented itself as the main resource investor so as to accommodate potential conflicts and to satisfy CSN friendly demands. It does not seem surprising that the foregoing remarkable changes in institutional arrangements reveal the crony relations between CSN and political entrepreneurs within the government sector. Provided that the federal government itself decided to assign to TLSA the construction (property) rights of the New Transnordestina Railway, it had submitted all the scheduled investments to government officials for their approval. The latter are the basis for defining financial parameters of the railway concession contract, such as the investment return rate. In addition, the public government became the ultimate financial backer of such a risky endeavor, since private investors were somehow put aside due to the changes in the contract arrangement. The encountered solution rewarded crony relations rather than exchange of property rights via the market. This is largely due to a traditional view of economic development tradition in Latin America that takes the government as playing a major role in promoting investment and therefore economic growth. On behalf of a politically appealing Latin American doctrine called National Developmentism, public officials pursue financing instruments to provide privileges to specific groups and businesses without public scrutiny. Unfortunately, such recurrent feature of the institutional matrix partly explains why Brazil has become an institutionalized rent-seeking society (Lisboa and Latiff 2013). In the specific case of Transnordestina railway, the federal and northeast state governments decided to present the investment project as a national priority to achieve regional development that would require large public financial resources. However, the final institutional arrangement between the government and CSN was very complex and prevented a clear specification of property rights, on which economic efficiency depends. A private company was entitled to manage construction of Transnordestina and its future cargo operation. Nonetheless, the Ministry of Transport would have to approve of any changes in budgetary issues, since different financial instruments could come from various types of public funds. Our hunch is that this setting sheds light on ill-defined property rights that in turn bring a grey area between the private and public rights and obligations related to accountability, compensation, assessment of socio-environmental impacts, and so forth. This is because, under the final contractual scheme, the conceding power and most of the capital are public and the manager is private without clear details about how exactly all the various involved players would share the responsibility for credibly completing Transnordestina railway. With the above in mind, we claim that the recurrent long investment delays and changes in budget schedules that turn out to be main features of Transnordestina project as a government failure are unintended consequences of the fragile institutional arrangement between federal and three Brazilian northeast state government officials and CSN.

17 3.Flawed contractual arrangements underlying Transnordestina project: Brazil s crony capitalism in the wild The solution, which seemed to tie the convergent interests between the private and the public sectors in 2006, would result in further problems. In our viewpoint, there is a problem of ill-defined rights of property which, in the perspective of Alchian and Demsetz (1972), entails management inefficiencies. The work of the New Transnordestina was considered one of the national priorities and got public resource injection for its execution (most of the paid-in capital, as seen in Section 1). However, the work management was in charge of a private company which, due to the legal solution found, would require authorization of the public sector (through the Ministry of Transport) for every increase in costs to review the budget, implying new capital injections of different entities and federal funds. There is a grey area between the private and public responsibility which contributes to the increase of the usual problems of dealing with large infrastructure works, such as compensation, fiduciary regularization, socio-environmental impacts etc. The conceding power is public, the manager is private, most of the capital is public, the private sector is responsible for the work but, due to the hybrid contractual scheme, the responsibility is imprecisely shared between public and private sectors. Therefore, the delays and reviews are mainly due to the legal-institutional arrangement settled between the Federal Government and the CSN: the Project Finance. If the PP or even a common concession were considered, it would be possible to elucidate the specific rights and obligations of each part involved, including the allocation of risks between them, ensuring the provisioning of guarantees, the governance appropriate tools and the criteria of budget review and terms. The analysis of the background of the Transnordestina Railway evidences that the adoption of a well-defined institutional mark would have avoided part of the problems posed in the work since the beginning. The construction of the Transnordestina Railway is regarded as a public policy of the State decision making, implemented by private agents converging a wide range of actors, interests and social ties. Besides the CSN and the Federal Government through the Ministry of Transport, other important actors in the process include: the Brazilian Institute of Environment and Renewable Natural Resources (IBAMA), which plays an important role in the environmental regulatory mark; the participations of the Investment Fund of the Northeast (FINOR), the National Bank of Economic and Social Development (BNDES) the Northeastern Bank (BNB), the Superintendent for the Development of the Northeast (SUDENE) and its Development Fund of the Northeast (FDNE), all of them relevant actors for the project finance. Other public power spheres were also involved in the work development, with consequences for the social, political and economic sectors. The Public Attorney and the Public Ministry of Labor acted to protect the workers and the population affected by the construction of the railway, and also the representatives of the subnational governments states and municipalities which had their interests positively or negatively affected by the work. Moreover, the conflicting scenario of the actors interaction in the search of solutions is important because it expressed the action of the government political stance to redefine the action of the State as an instrument to foment economic activities, mainly through the implementation of the Growth Acceleration Program (PAC), started during the President Luiz Inácio Lula da Silva`s term ( ).

18 For the Federal Government, the New Transnordestina Railway was part of a national project to insert the North and the Northeast in the Brazilian economy and the international market. The integration of the Ports of Suape and Pecém interlinking the Northeastern coast with the rest of Brazil, through the North-South Railway, opens a new route for the flow of primary goods minerals, grains from the new frontiers of production, and industrial goods, such as the industrial complex in Pernambuco (the backside area of the Industrial District of Suape) and Ceará (the backside area of the Port of Pecém). It is a structuring investment to try to meet one of the central issues in Brazil in the beginning of the 21st century. As pointed out by Fronzaglia et al. (2013), the historical aim of reducing the poverty and the secular regional inequality is recovered through a process of industrialization of the Northeastern area and the structuring of a logistic and transport network integrating the Northeast to the other Brazilian regions and the international economy. The difficulties of its implementation are expressed by the Engineering Clubs in the Northeast which, in a letter sent to former President Lula, recommended the recovery of the railway works by the state company VALEC and showed the obstacles to its execution. The TLSA is responsible for the execution, the delays in the schedule, the shape of railway and the exclusive management of its operation. We can say that the period between the concession of the Railway Network of the Northeast and the creation of the TLSA was marked by the convergence of the interests of the Federal Government and the CSN (a honeymoon which started to break up due to the conflicts resulting from the execution of the work, as depicted in Figure 1). Figure 1 Temporal scheme of the convergence and the divergence of the interests in the Transnordestina Railway work The project scheduled to begin in 2006 should have been completed in 2010 but it was extended to 2014 later on. The differences between the provisional and the effective costs,

19 the alterations of the project, and land, environmental and social issues caused the work to be completed in 2016, totaling ten years and not the original four-year schedule. In fact, with the expansion of the production of grains from the North of Tocantins, Piauí and other states of the North and Northeast, the opening of new fronts of production of minerals and, on the other hand, the expansion projects of the Ports of Suape and Pecém and the industrial production of the region (shipyards, Abreu Lima Oil Refinery, industrial hubs in Ceará, Pernambuco, etc.), the interests are articulated to guarantee the effective implementation of this outflow route. The claim posed in Section 2 was that the arrangement made for the execution of the work is in the origin of the current difficulties. The search then was for the easiest theoretically way. It was a shortcut to overcome the time spent in the complex preparation and the procedures inherent to projects of such magnitude. The Project Finance set forth provisions for a capital increase of the CSN, the TLSA s controller, resources from the Treasure and public financing, and established a partnership to reassign the responsibility of the construction to the beneficiary of the 1997 concession articulated with the governmental interests (see Figure 2). Figure 2 Temporal scheme of the convergence of interests Obstacles due to expropriations, environmental issues, social, land and political conflicts are important and inherent elements in any work of such magnitude but the model adopted and the conflicts among the several actors involved were potentialized and proved difficult to be overcome. The conflicts embodied the divergence of interests, as seen in Figure 3.

20 Figure 3 Scheme of the divergence of interests Interests, inseparable from the institutions, involve aims of each agent and actor involved in the process. At this point, the arrangement between the public and the private spheres becomes more clearly. The ideas comprise the collective images and representations, and the values expressed in the actors actions. They create the context and influence the actors behavior and also the configurations wherein the referential is constantly embedded and elaborated. According to Surel and Palier (2005), the joint mobilization of the three Is is based on the idea that the constituent and the explanatory variables of a public policy are not exclusive but are interwoven and associated so as to limit the dynamics of the actors action, something essential for understanding the State in action. The identification of the three Is and their effects in the case of the New Transnordestina Railway are depicted in Figure 4. Figure 4 The three Is in the case of the New Transnordestina Railway The Theory of the Three Is also integrates the conception of Crony Capitalism (LAZZARINI, 2011) insofar as they focus on the definition and the understanding of the dynamics of the arrangements between the public and private spheres and the interests of the State and the business groups. These bonds are associated with the set of social relations such as the ones involving the governmental decisions with merely economic interests.

21 It is worth pointing out then that contemporary capitalist activities are linked to form a network of properties, which also explain the creation of bonds due to the societarian participation of entrepreneurs, businessmen and companies, resulting in a set of projected connections among the parts which compose the capital of these companies. Lazzarini (2011) identifies a pattern in these connections, the agglomerations. In Brazil, these agglomerations do not only rely on private capital but also on the direct or indirect resources linked to the State, mainly the BNDES, the state companies and their pension funds. The participation of these actors enables an extensive and effective interconnection of the companies linking several agglomerations, creating an atmosphere for the convergence of certain business interests but also highlighting the conflicts. In this context, not only do the owners agglomerate themselves but the own companies are colligated in groups which have the same controller partners and often present mutually coordinated action strategies. These business agglomerations under a common controller cover are usually called economic groups. Examples of this logic include the TLSA, with the VALEC and the FNDE (run by the Superintendence of the Development of the Northeast- SUDENE) as main shareholders, and the CSN. Another important evidence to reinforce the bonds between the large companies is the participation of entrepreneurs and businessmen with seat in the administrative board of other companies. When these structures are evaluated, we can see a wide network of relations which corroborates the existence of true relations between the great capitals in Brazil, as shown in Figure 5, stressing the central position of the pension funds of the federal state companies.

22 Figure 5 Network of relations among the large corporations and pension funds in Brazil Source: Repórter Brasil (2014) 6 In the relations between public power/private agents (government/companies), the interconnected governmental actors and therefore the distinct agglomerations play an important role in business deliberations, mainly as far as investment decisions are concerned. This occurred more effectively from the privatizations under the President Fernando Henrique Cardoso s term on, and increased in Lula s government. As to Lazzarini (2011), the governmental actors actions in these networks are based on three elements: 1) the current Brazilian political system, including the political parties and the subjacent electoral process; 2) the relations between the political actors, the executive power, the state companies and their pension funds and 3) the coalitions among the great national business groups and the members of the political system. The main link among them is the campaign finance. Such relations can also be considered an extension of the deep-rooted patrimonialism in the Brazilian cultural politics (FAORO, 2001). Lazzarini (2011) also relates this event to the campaign donations showing that the companies which made the biggest political donations to elected candidates were those which got resources from state banks, mainly the BNDES. The relations of the CSN and later on the TLSA financial configurations are in this context. These elements form the politically dangerous liaisons between the public and private actors and the public and private interests, as schematically shown in Figure Available at < retrieved on 21/06/2014.

23 Political System INFLUENCE Government DONATIONS STRATEGIC ALIGNMENT RIGHTS OF CONTROL Private Companies CAPITAL Governmental Entities Figure 6 - The politically dangerous liaisons among the public and private actors Source: Lazzarini (2011, p. 55) The flow chart above shows only a unidirectional arrow between the private sector and the government as far as the strategic alignment is concerned. However, we can infer that a reverse arrow (government private enterprises) also expresses the reality of the Brazilian capitalist bonds. The configuration of the relations between the government and the large business groups also create situations wherein the government influences some corporate decisions, as in the case of the New Transnordestina Railway. 4.Implications and final remarks The analysis of the legal context, including the contract, legislation, interests, ideas and institutions of the New Institutional Economics, led us to the origin of this public policy: it is neither the result of a deep diagnosis nor the clear and objective demand of the affected subnational governments or the several private actors and sectors of society. The origin is in the networks of interests developed between the government and specific private agents, between the Labor Party government and one of its main supporters, the CSN. All the analysis provided here is an attempt to explain how such a singular action (or such an institutional legal deviation) is configured as public policy and has involved several actors, interests, values and references. It is where the public interest, in its clear republican meaning, fades out.